Medicare Data

点击翻译按钮获取中文摘要

In this project we will examine the temporal trajectories of Medicare Part A and B files, which contain records on individual episodes of service use, will be fully linked for the period 1982 to 2002 to the 1982, 1984, 1989, 1884, and 1999 NLTCS screening, community, and institutional interview records. Revisions of the Medicare computer files and data systems increased the information they provide about specific Medicare treatments and diagnoses over time. In part this is due to an improvement in the information content and structure of the databases and, in part, it is due to changes in medical practice and diagnostic precision. What is common to all years of Medicare data is that the exact dates, durations, and amounts of services consumed and expenditures made through Medicare are available on those files for each person in the NLTCS sample for each day in the entire time period 1982 to 2002. Thus, the Medicare records for individuals can be accumulated over the time, age, gender, and other demographic domains as needed to be related to the functional status and health changes manifest in the NLTCS interview data. The Medicare service use data for individuals will be examined for groups of individuals, both at a basic level for commonly (in the literature) used disability categories (e.g., 1-2, 3-4, 5-6 ADLs impaired), and by using objective scaling procedures to transform batteries of multiple survey measures into a small number of disability or health scores. Once the Medicare data are recorded over time and related to survey data, it will be possible to also look for evidence of period effects related to changes in the Medicare Program or in the availability of new types of therapies and diagnostic mechanisms for the elderly (65+) and oldest old (85+) populations and demographic subgroups.